BERT_top5_bm25_rr5_10_epoch

This model is a fine-tuned version of bert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.0294
• Accuracy: 0.7708
• F1: 0.6486
• Precision: 0.5385
• Recall: 0.8155

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 100
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
No log	0.2623	16	0.6133	0.8237	0.5270	0.8667	0.3786
No log	0.5246	32	0.5112	0.7758	0.2393	1.0	0.1359
No log	0.7869	48	0.4725	0.8363	0.6448	0.7375	0.5728
No log	1.0492	64	0.3894	0.8539	0.6882	0.7711	0.6214
No log	1.3115	80	0.7018	0.5013	0.5	0.3379	0.9612
No log	1.5738	96	0.4207	0.8338	0.7054	0.6529	0.7670
No log	1.8361	112	0.4159	0.7834	0.6587	0.5570	0.8058
No log	2.0984	128	0.4052	0.8060	0.6831	0.5929	0.8058
No log	2.3607	144	0.4456	0.7859	0.6743	0.5570	0.8544
No log	2.6230	160	0.3880	0.8564	0.7016	0.7614	0.6505
No log	2.8852	176	0.5137	0.8262	0.5660	0.8036	0.4369
No log	3.1475	192	0.4837	0.7935	0.6496	0.5802	0.7379
No log	3.4098	208	0.7301	0.7280	0.6197	0.4862	0.8544
No log	3.6721	224	0.6014	0.8413	0.6866	0.7041	0.6699
No log	3.9344	240	0.7912	0.7456	0.6481	0.5054	0.9029
No log	4.1967	256	0.6779	0.7834	0.6587	0.5570	0.8058
No log	4.4590	272	0.6352	0.8010	0.6749	0.5857	0.7961
No log	4.7213	288	0.9313	0.7229	0.6207	0.4813	0.8738
No log	4.9836	304	0.7459	0.7758	0.6454	0.5473	0.7864
No log	5.2459	320	0.6967	0.8186	0.6636	0.6396	0.6893
No log	5.5082	336	0.7340	0.8086	0.6780	0.6015	0.7767
No log	5.7705	352	0.9585	0.7506	0.6374	0.5118	0.8447
No log	6.0328	368	0.8556	0.8010	0.6749	0.5857	0.7961
No log	6.2951	384	1.0044	0.7758	0.6590	0.5443	0.8350
No log	6.5574	400	1.0174	0.7809	0.6641	0.5513	0.8350
No log	6.8197	416	0.8044	0.8111	0.6888	0.6014	0.8058
No log	7.0820	432	1.0973	0.7204	0.6159	0.4785	0.8641
No log	7.3443	448	0.9667	0.7758	0.6537	0.5455	0.8155
No log	7.6066	464	0.7502	0.8438	0.7130	0.6814	0.7476
No log	7.8689	480	1.0102	0.7733	0.6617	0.5399	0.8544
No log	8.1311	496	0.9457	0.7783	0.6589	0.5484	0.8252
0.2259	8.3934	512	0.9533	0.7834	0.656	0.5578	0.7961
0.2259	8.6557	528	1.0134	0.7783	0.6589	0.5484	0.8252
0.2259	8.9180	544	1.0594	0.7632	0.6466	0.5276	0.8350
0.2259	9.1803	560	1.0415	0.7708	0.6566	0.5370	0.8447
0.2259	9.4426	576	1.0485	0.7683	0.6515	0.5342	0.8350
0.2259	9.7049	592	1.0386	0.7708	0.6540	0.5375	0.8350
0.2259	9.9672	608	1.0294	0.7708	0.6486	0.5385	0.8155

Framework versions

• Transformers 4.40.1
• Pytorch 2.2.1+cu121
• Tokenizers 0.19.1